Getting the Optimal Service Efficiency That Autoscalers Won't Give You

Explore an innovative approach to optimizing Kubernetes microservices efficiency beyond standard autoscaling solutions in this 37-minute conference talk. Delve into the challenges of container sizing for CPU and memory in dynamic environments with frequent application changes and varying traffic patterns. Learn from a real-world case study of a business-critical financial services application for SMB customers, where the goal was to minimize cloud costs without compromising performance. Discover why configurations that minimized service costs were not recommended by autoscalers, which adjust resource sizing based on historical usage without considering actual cloud resource costs or application performance impact. Gain insights into an exploratory testing approach that achieves optimal service efficiency, surpassing the limitations of traditional autoscalers in Kubernetes environments.

Getting the Optimal Service Efficiency That Autoscalers Won’t Give You - Mauro Pessina, Moviri